I would say I’m a pretty competent smartphone user, yet I probably don’t use half of the apps or features. However, because I don’t know what I’m missing out on I remain unaware of any other functionality. With that said, if I could get more out of my phone without a steep learning curve or changing my habits at a user, I definitely wouldn’t say no!
This is why we’re so excited at SDL about the launch of SDL Trados Studio 2017 and our new transformational translation memory (TM) technology: upLIFT. This technology brings two key new features – upLIFT Fragment Recall and upLIFT Fuzzy Repair – which means that you can get much more out of your TMs than previously possible, without needing to change the way you work.
Let’s start with fragment recall: this aims at leveraging segment “fragments" (sometimes also referred to as subsegments) from the translation memory in a novel way, to automatically help translators work faster in both ‘no match’ and ‘fuzzy match’ scenarios. Let’s dig a bit deeper…
What is fragment recall and why use it?
Imagine you’re working on your translation and you have no TM match. In this scenario you have some options: you could perform a concordance search or enable machine translation. If you’re using Studio 2014 or 2015 you could even use an AutoSuggest Dictionary. However, all of these options require some sort of manual process and you can’t always be certain of the quality, or context, of the suggestion.
Enter fragment recall. This gives you intelligent fragment matches directly from your TM. In SDL Trados Studio 2017 there are two types matches:
- Matching fragments in new segments against whole translation units (TUs) indicated by this symbol:
- Matching fragments in new segments against corresponding fragments within existing TUs indicated by this symbol:
The difference can be seen in the screenshots below:
Figure 1: “No match" scenario with whole TU fragment recall
Figure 2: Matching fragments from translation units
When you have no segment match from your TM, these fragments are identified automatically and offered to you for use in the new ‘Fragment matches’ window, as well as through AutoSuggest as soon as you start to type.
The great thing about upLIFT Fragment Recall is that you can always see the context of the recalled fragment by clicking on the match. As you can see in figures 1 and 2, the match is highlighted in both the source and target segment in the match window, as well as the full context of the TU if the match comes from a TU fragment (figure 2), so you can be confident of its quality and that it’s appropriate for use in the current context.
What’s even better is that any new content that you add to your TM is automatically aligned and immediately available for fragment recall, saving you time and effort. No more manual dictionary creation or re-creation!
upLIFT Fragment Recall can also be used to extract terminology, as you can quickly add these matches to a termbase directly from the ‘Fragment matches’ window in Studio to help improve consistency across all your work.
With added batch processing functionality, you can even measure the impact of these fragment matches and report on it both pre- and post-translation.
Behind the scenes
To make upLIFT possible, behind the scenes we have changed the underlying TM technology. Finding full segment matches is much easier as it contains pairs of aligned segments, but this becomes more complex at subsegment level. Matching a TU fragment accurately (e.g. a phrase or term within a sentence) and retrieving the corresponding part of the translation requires a much finer level of alignment, which we call fine-grained alignment. upLIFT enables you to create new TMs with this new fine-grained alignment TM technology, or upgrade existing TMs so that you can reap the full benefits of this new feature.
It’s an exciting change in TM technology that really helps you get more out of your TM without extra effort or manual steps as you translate. Studio 2017 is designed to make the difference to your work and with upLIFT the difference is clear to see. Look out for our next blog on upLIFT Fuzzy Repair to learn more about this technology!